Cancer is a complex disease that kills millions of people annually. Alterations of genetic and epigenetic programs derail the normal controls that keep cells in homeostasis. The p53 tumor suppressor is crucial in maintaining homeostasis and its activity is lost in the development of most cancers. The most common alteration that occurs in the p53 pathway is a missense mutation of the p53 gene itself. Numerous in vivo and in vitro experiments suggest that cells with mutant p53 proteins haven additional growth advantages over cells that lack p53. Other studies show that tumor cell lines are addicted to mutant p53 expression as knockdown of mutant p53 reverts transformed phenotypes and makes cells more responsive to chemotherapy. Studies with human tissues and xenografts indicate that tumor cells evolve differently from the stroma and both contribute to the tumor phenotype. The stroma contains numerous components such as endothelial cells which form vessels to feed the tumor, immune cells that try to fight the cancer cells, and stromal fibroblasts that lay down the matrix and are often changing the microenvironment for a tumor cell to metastasize. Current mutant p53 mouse models generated do not faithfully recapitulate human sporadic tumors, either due to the presence of exogenous promoters driving expression of mutant p53 or the consequence of having a null allele (which has profound biological effects) to start with. To address these issues, we have generated the first somatic model of a p53 missense mutation via a knockin at the mutant p53 locus. This allele expresses wild type p53 but upon Cre-mediated recombination will delete the wild type cDNA and express mutant p53. This model allows us to make one single mutant p53 expressing cell in a sea of wild type cells and as such better models human cancer. We plan to generate somatic p53 mutations in breast epithelium of mammary prone tumor models. We will also make a p53 mutation in stromal fibroblasts in an ErbB2/neu model. In both, we will monitor changes to the microenvironment and ultimately perform next generation sequencing and expression analyses to understand both tumor and stromal specific changes that drive tumorigenesis. This model will provide an impetus for understanding the co-evolution of tumor epithelial cells and their microenvironment. Because our conditional mouse model shares the underlying molecular pathology with human sporadic tumors, it will be more predictive of human responses to drugs, and thus a more valuable tool in preclinical testing. Finally, to evaluate the potential therapeutic efficacy of inhibiting mutant p53 in cancers, we wil develop a novel conditional mutant p53 allele. In summary, through the generation of better mouse models, this work aims to advance our understanding of the mechanisms of how mutant p53 contributes to tumorigenesis and extend this knowledge to advance therapeutic options for tumors harboring these mutations.